HEALTH WARNING: THESE ATLASES ARE STILL IN DEVELOPMENT, AND COME
WITH NO WARRANTY WHATSOEVER
Experiment name:
Brain Imaging of Normal Subjects (BRAINS) age-specific MRI atlases from young adults to
the very elderly (v1.0)
Investigators:
David Alexander Dickie, Dominic E. Job, David Rodriguez, Andrew Robson, Samuel Danso,
Cyril Pernet, Mark E. Bastin, Ian J. Deary, Susan D. Shenkin, Joanna M. Wardlaw
Affiliations:
Edinburgh Imaging, The University of Edinburgh.
(http://www.ed.ac.uk/clinical-sciences/edinburgh-imaging)
(https://www.brainsimagebank.ac.uk/)
Correspondence: [email protected]
Description:
Atlas subjects and methods
We have developed seven age-specific atlases of T1 brain MRI from 25 to 92 years. The
number of subjects and age band of each atlas are summarised in the table below.
Group Name
Age (years)
N
Group00001
25-34
20
Group00002
35-44
24
Group00003
45-54
23
Group00004
55-64
13
Group00005
71-74
47
Group00006
75-78
50
Group00007
91-93
48
HEALTH WARNING: THESE ATLASES ARE STILL IN DEVELOPMENT, AND COME
WITH NO WARRANTY WHATSOEVER
The mean T1 intensity atlases for each age band were created by following a validated and
widely used protocol1. Briefly, within each age band,
1. All subjects were 6pt (rigid body) registered to a randomly selected subject
2. A T1 intensity average of the results from step 1 was taken
3. All subjects were 12pt registered to the average T1 image from step 2
4. A T1 intensity average of the results from step 3 was taken
5. All subjects were nonlinearly registered to the average T1 image from step 4
6. A T1 intensity average of the results from step 5 was taken
7. Steps 5 and 6 were repeated for a total of 20 further iterations (it).
This process is illustrated in the figure below.
Grey matter (GM), normal appearing white matter (WM), and cerebrospinal fluid (CSF)
tissue maps were estimated in the native space of each subject using T1 voxel neighbourhood
information and atlas based correction which may assist in successful segmentation of elderly
brains2, 3. T1 hypointensities in the white matter incorrectly classified as GM were removed
by defining a standard space region mask where GM almost universally does not occur, e.g.,
in the occipital caps of the lateral ventricles. This mask, shown in the figure below, was then
nonlinearly warped to each subject and applied to their GM mask. Probability maps for each
HEALTH WARNING: THESE ATLASES ARE STILL IN DEVELOPMENT, AND COME
WITH NO WARRANTY WHATSOEVER
tissue within each age band were estimated by 12pt registering each individual tissue mask to
the finally derived average T1 image within their age band.
The finally derived average T1 images from each age band are summarised in the figure
below.
Funding & Data sources
The 25-64 year data4 were acquired through NIH grant R01 EB004155-03 (PI: Dr Mark E.
Bastin), the 71-93 year data were acquired as part of the Lothian Birth Cohort5-7 (PIs: Prof
Ian J. Deary, Dr Mark E. Bastin, Prof John M. Starr, Prof Joanna M. Wardlaw) funded by a
Research into Ageing program grant, the Age UK funded Disconnected Mind project, and the
UK Medical Research Council. Brain Research Imaging Centre (BRIC), The University of
Edinburgh funded the creation of the templates.
HEALTH WARNING: THESE ATLASES ARE STILL IN DEVELOPMENT, AND COME
WITH NO WARRANTY WHATSOEVER
Citation guidelines
Use of these atlases in conference proceedings and journal publications should be cited as,
David Alexander Dickie, Dominic E. Job, David Rodriguez, Andrew Robson, Samuel Danso,
Cyril Pernet, Mark E. Bastin, Ian J. Deary, Susan D. Shenkin, Joanna M. Wardlaw (2016).
Brain Imaging of Normal Subjects (BRAINS) age-specific MRI atlases from young adults to
the very elderly. DOI: http://dx.doi.org/10.7488/ds/1369
Licensing
Creative Commons Attribution 4.0 International licence (CC-BY 4.0)
Bibliography
1.
2.
3.
4.
5.
6.
7.
Fonov V, Evans AC, Botteron K, Almli CR, McKinstry RC, Collins DL. Unbiased
average age-appropriate atlases for pediatric studies. Neuroimage. 2011;54:313-327
Avants BB, Epstein CL, Grossman M, Gee JC. Symmetric diffeomorphic image
registration with cross-correlation: Evaluating automated labeling of elderly and
neurodegenerative brain. Med. Image Anal. 2008;12:26-41
Zhang Y, Brady M, Smith S. Segmentation of brain mr images through a hidden
markov random field model and the expectation-maximization algorithm. IEEE
Trans. Med. Imaging. 2001;20:45-57
Dickie DA, Mikhael S, Job DE, Wardlaw JM, Laidlaw DH, Bastin ME. Permutation
and parametric tests for effect sizes in voxel-based morphometry of gray matter
volume in brain structural mri. Magn. Reson. Imaging. 2015;33:1299-1305
Deary I, Gow A, Taylor M, Corley J, Brett C, Wilson V, et al. The lothian birth cohort
1936: A study to examine influences on cognitive ageing from age 11 to age 70 and
beyond. BMC Geriatrics. 2007;7:doi:10.1186/1471-2318-1187-1128
Deary IJ, Gow AJ, Pattie A, Starr JM. Cohort profile: The lothian birth cohorts of
1921 and 1936. Int. J. Epidemiol. 2012;41:1576-1584
Wardlaw JM, Bastin ME, Valdés Hernández MC, Muñoz Maniega S, Royle NA,
Morris Z, et al. Brain aging, cognition in youth and old age and vascular disease in the
lothian birth cohort 1936: Rationale, design and methodology of the imaging protocol.
International Journal of Stroke. 2011;6:547-559